21–30 of 174 posts
Subreddit simulator is pretty convincing conversations, putting that to high quality voices? mannnn, so many good applications.
Speaking of which, why don't people just talk about the good applications. You'll get ostracized for speculating more bad things about COVID, but talk about how doomed we potentially are with deep fakes? Give that blogger a pulitzer prize!
How long does it take to generate a good quality voice?
I then transfer learned for each of these speakers. Some speakers have as little as 40 minutes of data, others have up to five hours. The resulting quality isn't strictly a function of the amount of training data, though more typically helps. It's also important to have high fidelity text transcriptions free of errors.
The transfer learning runs vary between six hours and thirty six hours.
I'm using 8xV100 instances to train glow-tts and 2x1080Ti to train melgan. I'm continuously training melgan in the background and simply adding more training data. The same model works for all speakers.
On a more positive side to this technology. I've been wondering about the possibility of using this sort of tech (or the API offerings from Azure or GCP) to provide voice overs in video games. By that I mean for smaller budget Indie development, it would be certainly interesting to either be able to generate voice audio from transcripts in order to add voices to background NPCs and so on (or even the possibility of d…
These companies tend to focus on off-the-shelf turnkey solutions, so they'll have a suite of a few voice actors to choose from for different character archetypes.
On a more positive side to this technology. I've been wondering about the possibility of using this sort of tech (or the API offerings from Azure or GCP) to provide voice overs in video games. By that I mean for smaller budget Indie development, it would be certainly interesting to either be able to generate voice audio from transcripts in order to add voices to background NPCs and so on (or even the possibility of d…
There are half a dozen startups in this space that provide the tech. They use embedded style tokens or sliders to change the emotion, pitch, timbre, etc. I don't have links off hand, but they're not too difficult to find. These companies tend to focus on off-the-shelf turnkey solutions, so they'll have a suite of a few voice actors to choose from for different character archetypes.
E.g. training off Schwarzenegger and offering an Arnold transform
On a more positive side to this technology. I've been wondering about the possibility of using this sort of tech (or the API offerings from Azure or GCP) to provide voice overs in video games. By that I mean for smaller budget Indie development, it would be certainly interesting to either be able to generate voice audio from transcripts in order to add voices to background NPCs and so on (or even the possibility of d…
That's super cool. I am worried about the potential abuse of this service, are there any existing services that can help to identify audio deep fakes like this one is for making them? Found Resemblyzer: https://github.com/resemble-ai/Resemblyzer
On a more positive note, when deepfakes become a problem, we will see the emergence of a culture where unsigned authoritative content is not paid any attention.
Earlier quoted context omitted.
There are half a dozen startups in this space that provide the tech. They use embedded style tokens or sliders to change the emotion, pitch, timbre, etc. I don't have links off hand, but they're not too difficult to find. These companies tend to focus on off-the-shelf turnkey solutions, so they'll have a suite of a few voice actors to choose from for different character archetypes.
Out of curiosity, are there legal concerns? E.g. training off Schwarzenegger and offering an Arnold transform
I believe (I'm not certain) that celebrity voice impersonation is legal as long as it is not used to sell or endorse a product.
Most models are trained on the original speaker's voice, but maybe only a little bit. Models might incorporate learning from many speakers. We might even be able to boil down a speaker representation to a small vector encoding in the future. It'll be interesting if we can capture the representation of a person with just a few numbers.
I don't think the legislature should be overly protective against machine learning. It seems obvious to me that neural networks will play a huge role in creating entirely virtual musicians and influencers. We're already seeing this start to happen. r9y9 on github has published some models that rival Vocaloid in lyrical ability.
At the same time, we don't want these techniques used to commit fraud, slander, or have them be used to falsely accuse someone of committing some act. These are things we might need new legal protections for.
But I don't know what I'm talking about. I'm not a lawyer.
really impressed by this.
If he thinks this is egregious, I'll take it down.
As an outlier not running javascript, I'm reaping what I sow, but it would be nice to me and others in the same boat if projects make their landing page viewable without the need for javascript.